You're not going to believe this: Inflation/deflation debate still alive?

[phirang]

Re: You're not going to believe this: Inflation/deflation debate still alive?

Just like the preposterous recruiting commercial!

But seriously... there are people whom I think know more about economics than I, who seem to take the deflationary scenario seriously. I am under the impression that the debate really boils down to what the government (and Fed) will do -- and how American consumers of credit will respond. I think EJ and the iTulip crowd have it right, but fundamentally this argument is more about handicapping government strategy and consumer response than it is about objective economics. I mean -- it isn't as though deflation is technically impossible. Rather, it is a question of how alert the Fed is to deflationary threats versus inflation (cue snide reference to helicopters), whether they have the ability to successfully halt a collapsing credit bubble and re-inflate (cue snide references to printing presses), and whether over-extended American consumers will borrow if even more credit is extended to them under easier terms (cue snide reference to deranged weasels).

You're completely correct: a friend of mine is doing his econ phd at MIT, and while he's not macro, he knows bernanke's work... and ultimately, it's the capricious voting masses that will drove the inflationary sword into their own bellies'.

Ceterius paribus and without politiks, without a doubt we'd enter a deflationary spiral. Yet, that tenacious political dimension persists and makes everything a fking 6-sigma event.

[FRED]

Re: You're not going to believe this: Inflation/deflation debate still alive?

Mish and Rick don't get this. And they don't get the fact that consumers and businesses don't need to borrow...the borrowing can be done by other actors.

Indeed, we've showed them this diagram and explained it to them before.

We're patient guys. We're happy to explain it again.

Money once borrowed into existence rarely disappears. Most money continuously changes form as it flows through the economy. As consumers and businesses reduce their borrowing of new money into existence, the government can increase its borrowing of new money into existence.

[bart]

Re: You're not going to believe this: Inflation/deflation debate still alive?

As consumers and businesses reduce their borrowing of new money into existence, the government can increase its borrowing of new money into existence.

And how "convenient" the cover of the drop in the System Open Market Account is for that eventuality.

[ASH]

Re: You're not going to believe this: Inflation/deflation debate still alive?

Some say the world will end in fire,
Some say in ice.
From what I’ve tasted of desire
I hold with those who favor fire.
But if it had to perish twice,
I think I know enough of hate
To know that for destruction ice
Is also great
And would suffice.

Robert Frost

You, sir, win the Galactic Institute's Prize for Extreme Cleverness! Bravo!

[zmas28]

Re: You're not going to believe this: Inflation/deflation debate still alive?

I'm not understanding this too well.
Lets say A lends B $100. B ends up with $100 cold hard cash, and $100 in hock in the ledger. A ends up with no change, since he gave up $100 to B but got back an IOU for $100 which security we assume he can mark to market at $100. Net system result is that we generate a security (IOU) worth $100 for a -$100 entry in B's ledger. The negative entry has no value as it cannot be traded. This increases the credit in the system by $100. This is System at state 1 (up $100 in tradeable assets).

If B now defaults on the loan and is unable or unwilling to pay, the only effect I see (other than the fact that A is screwed) is that the IOU now becomes worthless. This represents a deflationary change from state 1 at which time our monetary system was up by $100. The debt default was deflationary because the security became worthless.

If, instead B repays the loan amount to A, then A returns the IOU to B and we're back to where we started at time zero. Net result zero, which represents a deflationary change from state 1.

So terminating the loan and therefore the security is deflationary either way. The green stuff doesn't die unless someone retires it; it can grow by printing more. The virtual stuff (credit) can grow or die. Yess-no?

[phirang]

Re: You're not going to believe this: Inflation/deflation debate still alive?

from my econ friend at MIT:

Empirical facts:
1. define seignorage as revenue from money creation. historically hyper inflation -- has it been due to bubbles or money growth policies which don't maximize seignorage?
A: if you take the perspective of adaptive expectations, you will find that it is due to bad mismanaged money growth policies. if you take the perspective of rational expectations, bubble theories can explain it. personally -- i am inclined to think adaptive expectations paints a more accurate picture of the mechanism. i think most neuro research and most psych+econ research suggests that people operate with adaptive type expectatiosn.
2. real variables are pro-cyclical (consumption, investment vary with output). here by cyclical i mean the phase is in sync with gdp or output.
3. real wages are acyclical.
4. exportation is acyclical -- the us is surprisingly closed despite what people like to thikn.
5. government spending is acyclical.
6. employment procyclical but lagged -- so movement in output followed by movement in employment.
7. productivity procyclical but in terms of ability explain output, productivity becomes a very small component.
8. real interest rate is slightly counter cyclical.
9. nominal interest rate is slightly pro cyclical.
10. money is procyclical, both nominal and real money. this also means that in the short run M can move with Y and M/P can move with Y, meaning we can think of P not responding one for one with Y. (suggests rigidity in P, where P is some average price measure.)
11. firms are not perfectly competitive.
Why do I list off this? Because any framework or language to discuss macro questions actually needs to be able to account for all of this behavior. The way most people talk about typical macro issues, they look at things in isolation. They have a theory of inflation in their head or a theory of employment. But it isn't clear that when you put these isolated narratives together, they actually can co-exist in a macro-framework story!
Framework stuff:
1. stories without nominal rigidities do not allow us to say that money is neutral or super neutral. this means that unless we assume a sort of nominal rigidity (general commodity prices or wage prices) we won't have real effects of money. but our stylized facts suggest otherwise.
2. because no1 seriously thinks there is perfect competition, monopolistic competition is a much better way to look at things. the nice thing is that this framework treats firms as seeing a market average price, and pricing their marginally differentiated good accordingly. this actually is a good language to discuss rigidities.
3. we want to model the mechanism of the rigidity. there are a number of different stories you could tell. you could think about at each t, some group of firms seek to reevaluate their prices, p_it-1, and make it into p'_it. so now at t, the market average price P_t-1 changes to P'_t. the way to spice up the story -- (a) firms only respond in price to sufficient changes in their own cost above a threshold -- they don't change/respond to differential movements, (b) menu costs, (c) mis-pricing theories, (d) timed price contracts -- think about labor-wage contracts that go on for months or years that do not respond to things like changes in CPI or other factors. the point is there are a LOT of (fairly convincing) stories to be told why firms aren't magically updating their prices perfectly. this is one simplified story i am giving you, but you can do much more intricate narratives. the point is you get very similar results.
the result from these stories is: (a) price stickiness, (b) not (much) inflation stickiness. observe that these frameworks still adhere to a rational expectation. this suggests we need to...
4. modify the narrative a bit by including adaptive expectations. people overweight recent information relative to other past information.
Conclusions to be drawn:
1. loosely speaking you have 3 margins to discuss:
(A) the output today is increasing in expected future output, and decreasing in the real interest rate. (this means that output today is decreasing in the nominal interest rate and increasing in expected inflation). this makes sense right? if the real interest rate is really high, you want to save more. or alternatively, if inflation is going to be super high tomorrow, you might as well just build your products today and sell today right?
(B) the nominal interest rate is decreasing in money balance, increasing in price, increasing in output. again this should pass the intuition of the smell test.
(C) inflation today is not only an increasing function of our deviation from optimum (i.e. our output without rigidities minus output due to rigidities, shocks), but also of future expected inflation, and with adaptive expectations a bit due to previous inflation. so an oil shock, think about it as a productivity shock, actually decreases the output due to rigidities, shocks, and therefore increases inflation today. the actual mechanism of this is due to the micro individual and firm behavior story we told before.
by the way, why do we talk about this difference between output w/o rigidities and output with rigidities. well if you believe that efficiency is a good normative outcome, w/o rigidities, in a market w/ monopolistic competition, this outcome is pareto optimal. let us call it the second best outcome. it is second best is b/c first best would be if we had perfect competition, but unfortunately we do not. so the way to understand it is that rigidities, shocks -- they move us away from second best which means that it is welfare harming. so our policies should be reactionary to move us back to second best to be welfare improving. (this is the intuition.)
now this is a very trivialized story and in practice no1 uses this per se. it is a pedagogical tool. but for conversation purposes, the reason it is a good thing to have in mind is because it gives you a coherent narrative of the macro economy that does fairly well to systematically explain all the macro-metric facts. in practice what is used is called Dynamic Stochastic General Equilibrium.
by the way it is worthwhile to mention that this stuff is actually used in practice. now i dont know if people are aware of this at the everyday level on wall st. because what happens is that it is usually contracted or some really hardcore quant phd is hired to run these things. and then the general conclusions are presented more in the style that i talk of above. but the pt is any real macrometric analysis that is done nowdays is only through DSGE. the Fed uses DSGE, the IMF, and more importantly every consulting company for businesses, they hire lots of macro-metric guys to run DSGE for them. so i imagine the macro-metric guys don't get the big bucks for nothing, otherwise mckinsey is just being dumb, no? many (most?) hedge funds also seem have quants that run DSGE somewhere. i think the problem w/ DSGE is that it is pedagogically and conversationally unattractive cuz of its complexity. so basically if you have intuition X, just ask a quant to run a DSGE to see if the conclusion is close to X. if so, just present your intuition X in the meeting and then move on, instead of using the DSGE, you know?
2. ok so let us look at the quote there.
my big issue is up top where he says basically people borrow "too" much relative to economic growth. specifically he says "do not understand that inflation is created by borrowing money. That is how money is created. Through borrowing. Inflation happens when people borrow at a rate that exceeds the rate of economic growth."
what does that mean? well, let's write a simple linearized model so i can talk about variables:
(A) y = Ey - a(i-Epi) = Ey - ar
(B) m-p=by-ci
(C) pi=Epi+d(y - y_sb) where y_sb is the second best output.
that is today's output is a function of expected future output and decreasing in real interest rate. why is it linear? well if you had a multi-dimensional model in which people were making decisions about things through time, you will have some equilibrium path. because firms and people are trying to do the best they can (i.e. optimizing), their decisions actually lead to a welfare optimum point. so what we care about is when the system moves away from the welfare optimum-- that is, deviations from equilibrium path. therefore you can log-linearize the system around a steady state and think of the linear equations above as log deviations from steady state.
what does people borrowing too much mean. intuitively this means that people today are demanding too much, higher y, and therefore in (C) the y-y_sb term is too high, i.e. non-zero, and therefore pi is higher. that is fine. except when you think about it, people borrow whatever they want given interest rates. that is actually the equilibrium outcome. this model talks about deviation from trend. so we can even think of the inflation here as deviation from trend inflation. the type of inflation your friend is talking about isn't the central "bad" inflation. there is, if you will, some regularized or trend notion of inflation that will occur, just like you will have some regularized or trend notion of growth, etc. where things are "bad" is when we get pushed out of an efficient equilibrium -- deviations from trend. people borrowing is an optimizing behavior in response to some price/interest rate scheme. so there isn't anything out of equilibrium with that unless you give a neuroscience style story of time inconsistent consumer myopia. (now i do research on that and i'm inclined to listen to such stories. but this isn't the position your friend is offering.)
now maybe what your friend wants to say is something like there is an adverse productivity shock which means y_sb (the natural output) is actually lower now because productivity is lower. and say we expect this to be persistent, so y_sb tomorrow, day after, all are lower. in turn people don't want to save and want to eat more today by borrowing. this means that the y-y_sb term is lower and therefore by (C) pi is higher. but observe that it wasn't borrowing or "over-borrowing" that was the exogenous culprit of the inflation increase. it was actually a productivity shock that bucked us off trend. over-borrowing was only the mechanism through which people could actually eat too much today.
my more broad point is that i think he doesn't frame thinking about inflation problems the right way. it is important to frame things correctly to draw the right conclusions. look some of his descriptions, mechanistically, sure, they are correct. but they do not at all tackle the deeper inflation problems. and that doesn't help us because it doesn't address what we could do to fix the problem.
yeah people shouldn't make snapshot judgments of inflation by wage prices perhaps. (does he mean real or nominal by the way?!) but a deeper reason for why is because wage prices themselves could have lots of rigidities. they are certainly not rapid moving creatures. and the way you tell your rigidities stories, that fully informs how you use wage data and price data to get a sense of the inflation!
i hope that helps ... it was probably annoyingly long ... anyway i am not a macro guy. i have a lot of methodological and theoretical issues with macro. that is for another day though =) ...

[Guest]

Re: You're not going to believe this: Inflation/deflation debate still alive?

Rick Ackerman is a very smart, persuasive and eloquent commentator. He's a decent guy too. But he has his head firmly tucked under the blankets in 2008. What part of this data does not at least inspire his professional curiosity?

The following table lists year-over-year inflation as of June 2008

[FRED]

Re: You're not going to believe this: Inflation/deflation debate still alive?

Interesting comments. Not clear who is the "he" your friend is referring to?

[phirang]

Re: You're not going to believe this: Inflation/deflation debate still alive?

Interesting comments. Not clear who is the "he" your friend is referring to?

"firiend"

I think the two big mistakes that people make is to judge inflation by wage prices, and to fail to understand the debt default is not deflationary.
First, Rick and Mish do not understand that inflation is created by borrowing money. That is how money is created. Through borrowing. Inflation happens when people borrow at a rate that exceeds the rate of economic growth.
Second, folks like Rick and Mish believe deflation happens through defaulted loans. Their argument is that widespread defaults lead to deflation.
But when loans are written off through default, there is no deflation because the money that was borrowed into existence was already spent and remains in the economy.
Only paying back loans results in deflation, because only paying a loan back removes that money from the economy. Paying back loans is flipside of inflation. Defaults are not the flipside of inflation. You can have widespread defaults and still have high inflation, but you cannot have people paying back their loans without a falling money supply and deflation.

[MLM]

Re: You're not going to believe this: Inflation/deflation debate still alive?

A significant difference between Argentina and the U.S. is the percentage of the global economy the U.S. represents. Argentina had a large "heat sink" to work against. The U.S. is big enough as a consumer that knock-on effects in the rest of the world will be significant.

[ASH]

Re: You're not going to believe this: Inflation/deflation debate still alive?

I'm not understanding this too well.
Lets say A lends B $100. B ends up with $100 cold hard cash, and $100 in hock in the ledger. A ends up with no change, since he gave up $100 to B but got back an IOU for $100 which security we assume he can mark to market at $100. Net system result is that we generate a security (IOU) worth $100 for a -$100 entry in B's ledger. The negative entry has no value as it cannot be traded. This increases the credit in the system by $100. This is System at state 1 (up $100 in tradeable assets).

If B now defaults on the loan and is unable or unwilling to pay, the only effect I see (other than the fact that A is screwed) is that the IOU now becomes worthless. This represents a deflationary change from state 1 at which time our monetary system was up by $100. The debt default was deflationary because the security became worthless.

If, instead B repays the loan amount to A, then A returns the IOU to B and we're back to where we started at time zero. Net result zero, which represents a deflationary change from state 1.

So terminating the loan and therefore the security is deflationary either way. The green stuff doesn't die unless someone retires it; it can grow by printing more. The virtual stuff (credit) can grow or die. Yess-no?

Hi zmas. I think the basic issue is that A is a bank, and when A makes a loan, it generates credit rather than "giving up" money. It isn't loaning physical money from its vault. The ability of A to create credit is only constrained by its reserve fraction requirement. When it creates credit, the deposit that is presented to the borrower doesn't come out of the bank's reserves -- rather, it is creating something that theoretically COULD be redeemed from the reserves if one so chose. As long as that credit is convertible into "money", it serves as money itself. That is why credit creation effectively adds to the money supply. The reserve fraction requirement sets a limit on how much credit a bank is allowed to create against its reserves. If everyone tried to collect from the reserves at once, then the bank would be screwed, because it ain't 1:1. In fact, one of the best articles in the iTulip archives is What (really) happened in 1995? by Aaron Krowne. It explains that the reserve fraction requirement was actually abolished for certain types of accounts.

So, from the standpoint of the money supply, in state 1, A has a security to market and B has a deposit at A to spend. A does not in fact have a balancing reduction in its reserves. Rather, its ability to create further credit against its reserves has been reduced by a fraction of the value of the loan.

If you now accept that in state 1 we're up by $200 instead of $100, then we're on the same page.

DISCLAIMER: I ain't no expert. This is how I understand credit creation and the banking system, but it is possible that I'm laboring under a massive misconception. I hope that the more knowledgeable members of the community will correct me if I have erred.

[zmas28]

Re: You're not going to believe this: Inflation/deflation debate still alive?

Hi zmas. I think the basic issue is that A is a bank, and when A makes a loan, it generates credit rather than "giving up" money. It isn't loaning physical money from its vault. The ability of A to create credit is only constrained by its reserve fraction requirement. When it creates credit, the deposit that is presented to the borrower doesn't come out of the bank's reserves -- rather, it is creating something that theoretically COULD be redeemed from the reserves if one so chose. As long as that credit is convertible into "money", it serves as money itself. That is why credit creation effectively adds to the money supply. The reserve fraction requirement sets a limit on how much credit a bank is allowed to create against its reserves. If everyone tried to collect from the reserves at once, then the bank would be screwed, because it ain't 1:1. In fact, one of the best articles in the iTulip archives is What (really) happened in 1995? by Aaron Krowne. It explains that the reserve fraction requirement was actually abolished for certain types of accounts.

So, from the standpoint of the money supply, in state 1, A has a security to market and B has a deposit at A to spend. A does not in fact have a balancing reduction in its reserves. Rather, its ability to create further credit against its reserves has been reduced by a fraction of the value of the loan.

If you now accept that in state 1 we're up by $200 instead of $100, then we're on the same page.

DISCLAIMER: I ain't no expert. This is how I understand credit creation and the banking system, but it is possible that I'm laboring under a massive misconception. I hope that the more knowledgeable members of the community will correct me if I have erred.

Ash, thanks for an interesting discussion. I agree that a bank can create money because its reserve ratio is less than 1. So at state 1 in this case (when A is a bank), the system has added more than $100 but (I think) a little less than $200, the exact amount depending on the reserve requirement (here I'm thinking that, on average, the bank would be required to retain a fraction of the deposits).
But my thought is that debt termination by either default or through repayment is, in fact, deflationary because it destroys the debt security.

[Charles Mackay]

Re: You're not going to believe this: Inflation/deflation debate still alive?

Defaulting on a bank loan is not the only type of default. If my portfolio consists of GM stock and GM bonds and that company goes broke, my wealth disappears.. that is deflationary for me personally. If that happens to a hundred million people it's deflationary for the country as a whole. You can argue that that money was spent by GM "and is still out there" but it's in the form of worthless buildings, equipment, and now unemployed workers.

[ASH]

Re: You're not going to believe this: Inflation/deflation debate still alive?

But my thought is that debt termination by either default or through repayment is, in fact, deflationary because it destroys the debt security.

I'm in agreement with you there. It's a question of what your basis is. My understanding is that those who are arguing that loan defaults "don't reduce the money supply" mean relative to the basis before the loan was made. Obviously, destruction of the debt security DOES reduce the money supply relative to the basis after the loan was made.

So, in the simple analysis, a loan default doesn't return the money supply to the state it was in before the loan was made because the credit is still out there. However, as I pointed out earlier, if a bank was counting upon a pile of bad debt securities to be part of its reserves (perhaps indirectly), then it will choose to reduce its portfolio of loans -- and the resulting reduction in the money supply will be leveraged by the reserve fraction. That mechanism could potentially reduce the money supply by more than the amount created by the original loan -- quite deflationary.

So much for the debt securities... what can wipe out the credit created by the loan? Bank defaults. The extra money relative to the basis before the loan is in the form of bank credit. Once those credits can no longer be exchanged for money (because the bank folds) they cease to function as money.

I also agree with Charles Mackay's observation about other types of asset deflation, although I suspect an economist (which I'm obviously not) -- or EJ -- might prefer to draw a clear distinction between changes to the money supply versus general asset price deflation. I think the point of bubble sites is that one is expecting asset price deflation; about the only debate we could be having is over the ultimate impact on the money supply. Right now we're seeing asset price deflation of houses concurrent with rising prices for gasoline. I thought the inflation/deflation debate was over whether or not we'd continue to see consumer prices rise, with both parties agreeing that the price of bubble assets would fall.

Again, the way I read the official party line, no one should be claiming that deflation is impossible. If deflation following the collapse of a real estate bubble were impossible, then Japan's lost decade (or is that lost two decades?) wouldn't have happened. Rather, iTulip has been saying that we'll get inflation as the result of Fed policy designed to head off the threat of deflation.

[Charles Mackay]

Re: You're not going to believe this: Inflation/deflation debate still alive?

I also agree with Charles Mackay's observation about other types of asset deflation, although I suspect an economist (which I'm obviously not) -- or EJ -- might prefer to draw a clear distinction between changes to the money supply versus general asset price deflation.

I was just musing that a bond market default is potentially a much bigger problem than bank loan defaults. Is a bond default considered asset deflation and a bank loan default a money supply deflation? ... I'll leave that to others. But as far as my investment stance is concerned, I'm betting that inflation will win out. I AM extremely worried about a stock and bond market crash set off by a scary dollar sell off this summer/fall. It's a Fibonacci 21 years since that exact scenario played out in 1987 and that happened when we were in a bull market. Now we are in a bear market (in real terms) and the outcome could be quite different...particularly when combined with the real estate crash that is in progress. And with California's median home price down 30% April 2008 over April 2007 it's certainly prudent to start labeling it a crash.